| dbo:abstract |
変分メッセージパッシング（へんぶんメッセージパッシング、英語: Variational message passing、VMP）はJohn Winnによって開発された、指数族のを用いた離散、連続ベイジアンネットワークを近似的に推論するための手法である。VMPはLatent Dirichlet allocation（LDA）などの手法で利用されるを一般化した手法であり、各々のノードの周辺分布を、その上に存在するメッセージを用いて逐次的に更新し、その近似解を求める。 (ja) Variational message passing (VMP) is an approximate inference technique for continuous- or discrete-valued Bayesian networks, with conjugate-exponential parents, developed by John Winn. VMP was developed as a means of generalizing the approximate variational methods used by such techniques as latent Dirichlet allocation, and works by updating an approximate distribution at each node through messages in the node's Markov blanket. (en) |
| dbo:wikiPageExternalLink |
http://www.johnwinn.org/Publications/papers/VMP2004.pdf http://dimple.probprog.org http://vibes.sourceforge.net http://research.microsoft.com/infernet https://web.archive.org/web/20050428173705/http:/www.cs.toronto.edu/~beal/thesis/beal03.pdf http://www.cs.toronto.edu/~beal/thesis/beal03.pdf |
| dbo:wikiPageID |
2461612 (xsd:integer) |
| dbo:wikiPageLength |
5835 (xsd:nonNegativeInteger) |
| dbo:wikiPageRevisionID |
1103078029 (xsd:integer) |
| dbo:wikiPageWikiLink |
dbc:Bayesian_networks dbr:Normalization_factor dbr:Approximate_inference dbr:Relative_entropy dbr:Mean dbr:Gamma_distribution dbr:Conjugate_prior dbr:Latent_Dirichlet_allocation dbr:Graphical_models dbr:Exponential_distribution dbr:Exponential_family dbr:Bayesian_networks dbr:Dirichlet_distribution dbr:Gaussian_distribution dbr:Poisson_distribution dbr:Markov_blanket dbr:Variational_Bayesian_methods dbr:Sufficient_statistic dbr:Natural_parameter dbr:Conjugate_exponents |
| dbp:wikiPageUsesTemplate |
dbt:Cite_journal dbt:Cite_thesis dbt:Main dbt:Reflist |
| dcterms:subject |
dbc:Bayesian_networks |
| gold:hypernym |
dbr:Technique |
| rdf:type |
dbo:TopicalConcept yago:WikicatBayesianNetworks yago:Abstraction100002137 yago:Group100031264 yago:Network108434259 yago:System108435388 |
| rdfs:comment |
変分メッセージパッシング（へんぶんメッセージパッシング、英語: Variational message passing、VMP）はJohn Winnによって開発された、指数族のを用いた離散、連続ベイジアンネットワークを近似的に推論するための手法である。VMPはLatent Dirichlet allocation（LDA）などの手法で利用されるを一般化した手法であり、各々のノードの周辺分布を、その上に存在するメッセージを用いて逐次的に更新し、その近似解を求める。 (ja) Variational message passing (VMP) is an approximate inference technique for continuous- or discrete-valued Bayesian networks, with conjugate-exponential parents, developed by John Winn. VMP was developed as a means of generalizing the approximate variational methods used by such techniques as latent Dirichlet allocation, and works by updating an approximate distribution at each node through messages in the node's Markov blanket. (en) |
| rdfs:label |
変分メッセージパッシング (ja) Variational message passing (en) |
| owl:sameAs |
freebase:Variational message passing yago-res:Variational message passing wikidata:Variational message passing dbpedia-ja:Variational message passing https://global.dbpedia.org/id/4xWif |
| prov:wasDerivedFrom |
wikipedia-en:Variational_message_passing?oldid=1103078029&ns=0 |
| foaf:isPrimaryTopicOf |
wikipedia-en:Variational_message_passing |
| is dbo:wikiPageDisambiguates of |
dbr:Message_passing_(disambiguation) dbr:Passing dbr:VMP |
| is dbo:wikiPageWikiLink of |
dbr:Message_passing_(disambiguation) dbr:One-shot_learning dbr:Approximate_inference dbr:Passing dbr:VMP dbr:Variational_Bayesian_methods dbr:List_of_statistics_articles dbr:Outline_of_machine_learning |
| is foaf:primaryTopic of |
wikipedia-en:Variational_message_passing |